Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,9 @@ New modules

* `Data.List.Relation.Binary.Permutation.Declarative{.Properties}` for the least congruence on `List` making `_++_` commutative, and its equivalence with the `Setoid` definition.

* `Relation.Binary.Construct.Add.Point.Order` to extend a given (order) relation so that
the point is below everything else in `Pointed A`.

Additions to existing modules
-----------------------------

Expand Down Expand Up @@ -99,6 +102,11 @@ Additions to existing modules
updateAt (padRight m≤n x xs) (inject≤ i m≤n) f ≡ padRight m≤n x (updateAt xs i f)
```

* In `Relation.Binary.Definitions`
```agda
Directed _≤_ = ∀ x y → ∃[ z ] x ≤ z × y ≤ z
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why binary? Don't you want to say that for any I-indexed family of points, there's a 'z' that is below all of them?

Copy link
Contributor Author

@jamesmckinna jamesmckinna Aug 23, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

See below.

But indeed, generalising may also be worthwhile, but ... downstream?

```

* In `Relation.Nullary.Negation.Core`
```agda
¬¬-η : A → ¬ ¬ A
Expand Down
109 changes: 109 additions & 0 deletions src/Relation/Binary/Construct/Add/Point/Order.agda
Original file line number Diff line number Diff line change
@@ -0,0 +1,109 @@
------------------------------------------------------------------------
-- The Agda standard library
--
-- A pointwise lifting of a relation to incorporate an additional point,
-- assumed to be 'below' everything else in `Pointed A`.
------------------------------------------------------------------------

{-# OPTIONS --cubical-compatible --safe #-}

-- This module is designed to be used with
-- Relation.Nullary.Construct.Add.Point

open import Relation.Binary.Core using (Rel; _⇒_)

module Relation.Binary.Construct.Add.Point.Order
{a ℓ} {A : Set a} (_≲_ : Rel A ℓ) where

open import Data.Product.Base using (_,_)
open import Level using (Level; _⊔_)
open import Function.Base using (id; _∘_; _∘′_)
import Relation.Binary.Construct.Add.Point.Equality as Equality
open import Relation.Binary.Structures
using (IsPreorder; IsPartialOrder)
open import Relation.Binary.Definitions
using (Reflexive; Transitive; Antisymmetric; Directed; Decidable; Irrelevant)
import Relation.Binary.PropositionalEquality.Core as ≡
open import Relation.Nullary.Decidable.Core as Dec
using (yes; no)
open import Relation.Nullary.Construct.Add.Point as Point
using (Pointed; ∙ ;[_])


private
variable
ℓ′ : Level
x∙ : Pointed A
x y : A

------------------------------------------------------------------------
-- Definition

infix 4 _≲∙_

data _≲∙_ : Rel (Pointed A) (a ⊔ ℓ) where
∙≲_ : ∀ x∙ → ∙ ≲∙ x∙
[_] : x ≲ y → [ x ] ≲∙ [ y ]

pattern ∙≲∙ = ∙≲ ∙

------------------------------------------------------------------------
-- Relational properties

[≲]-injective : [ x ] ≲∙ [ y ] → x ≲ y
[≲]-injective [ x≲y ] = x≲y

≲∙-refl : Reflexive _≲_ → Reflexive _≲∙_
≲∙-refl ≲-refl {∙} = ∙≲∙
≲∙-refl ≲-refl {[ x ]} = [ ≲-refl ]

≲∙-trans : Transitive _≲_ → Transitive _≲∙_
≲∙-trans ≲-trans (∙≲ _) _ = ∙≲ _
≲∙-trans ≲-trans [ x≲y ] [ y≲z ] = [ ≲-trans x≲y y≲z ]

≲∙-directed : Directed _≲_ → Directed _≲∙_
≲∙-directed dir ∙ ∙ = ∙ , ∙≲∙ , ∙≲∙
≲∙-directed dir [ x ] ∙ = let z , x≲z , _ = dir x x in [ z ] , [ x≲z ] , (∙≲ _)
≲∙-directed dir ∙ [ y ] = let z , _ , y≲z = dir y y in [ z ] , (∙≲ _) , [ y≲z ]
≲∙-directed dir [ x ] [ y ] = let z , x≲z , y≲z = dir x y in [ z ] , [ x≲z ] , [ y≲z ]

≲∙-dec : Decidable _≲_ → Decidable _≲∙_
≲∙-dec _≟_ ∙ _ = yes (∙≲ _)
≲∙-dec _≟_ [ x ] ∙ = no λ()
≲∙-dec _≟_ [ x ] [ y ] = Dec.map′ [_] [≲]-injective (x ≟ y)

≲∙-irrelevant : Irrelevant _≲_ → Irrelevant _≲∙_
≲∙-irrelevant ≲-irr (∙≲ _) (∙≲ _) = ≡.refl
≲∙-irrelevant ≲-irr [ p ] [ q ] = ≡.cong _ (≲-irr p q)

------------------------------------------------------------------------
-- Relativised to a putative `Setoid`

module _ {_≈_ : Rel A ℓ′} where

open Equality _≈_

≲∙-reflexive : (_≈_ ⇒ _≲_) → (_≈∙_ ⇒ _≲∙_)
≲∙-reflexive ≲-reflexive ∙≈∙ = ∙≲∙
≲∙-reflexive ≲-reflexive [ x≈y ] = [ ≲-reflexive x≈y ]

≲∙-antisym : Antisymmetric _≈_ _≲_ → Antisymmetric _≈∙_ _≲∙_
≲∙-antisym ≲-antisym ∙≲∙ ∙≲∙ = ∙≈∙
≲∙-antisym ≲-antisym [ x≤y ] [ y≤x ] = [ ≲-antisym x≤y y≤x ]

------------------------------------------------------------------------
-- Structures

≲∙-isPreorder : IsPreorder _≈_ _≲_ → IsPreorder _≈∙_ _≲∙_
≲∙-isPreorder ≲-isPreorder = record
{ isEquivalence = Equality.≈∙-isEquivalence _ isEquivalence
; reflexive = ≲∙-reflexive reflexive
; trans = ≲∙-trans trans
} where open IsPreorder ≲-isPreorder


≲∙-isPartialOrder : IsPartialOrder _≈_ _≲_ → IsPartialOrder _≈∙_ _≲∙_
≲∙-isPartialOrder ≲-isPartialOrder = record
{ isPreorder = ≲∙-isPreorder isPreorder
; antisym = ≲∙-antisym antisym
} where open IsPartialOrder ≲-isPartialOrder
5 changes: 5 additions & 0 deletions src/Relation/Binary/Definitions.agda
Original file line number Diff line number Diff line change
Expand Up @@ -96,6 +96,11 @@ Asymmetric _<_ = ∀ {x y} → x < y → ¬ (y < x)
Dense : Rel A ℓ → Set _
Dense _<_ = ∀ {x y} → x < y → ∃[ z ] x < z × z < y

-- Directedness (but: we drop the inhabitedness condition)

Directed : Rel A ℓ → Set _
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you give a reference for this definition? Google did not help me find anything relevant.

Copy link
Contributor Author

@jamesmckinna jamesmckinna Aug 23, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

https://en.wikipedia.org/wiki/Directed_set

The definition is taken from #2809 where it is currently called SemiDirected, but/and I'm not sure the Semi really makes sense. Moreover its use there can be (better?) refactored into this one, plus a use of change-of-base via _on_. So it definitely seems worth adding on its own terms, in some form or other.

The official definition requires A also to be inhabited (which can be finessed in any mode-of-use by an additional assumption x : A), but the 'condition' is indeed this one of having binary (and hence: any finite) upper bounds.

The lemma ≲∙-directed is precisely motivated by the observation that any relation satisfying the condition may be freely completed (preserving and reflecting the existing instances) to an inhabited relation satisfying the condition. It is the core of the 'lifting' construction on (pre)domains, but is minimal wrt its commitments to any other properties of the underlying relation. Not finding such a lemma motivated this PR as an addition/'infrastructure'...

But it perhaps/probably makes more sense to uncouple the definition of Relation.Binary.Construct.Add.Point.Order from these considerations, until we agree on suitable names/definitions for 'directed'ness as a property?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks - and @TOTBWF also expanded on this. There's a fairly non-trivial refactor of that PR incoming, after we discussed how to make things more stdlib-friendly.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

See my comments on #2829 ... I (still) think that we should isolate, and agree upon, a definition of Directed, and then use it appropriately downstream, as here, or in whatever version of DCPO we end up adopting...

Copy link
Contributor Author

@jamesmckinna jamesmckinna Sep 18, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmmm... if we instead go for

FinitelyDirected : Rel A ℓ  Set _
FinitelyDirected _≤_ =  {n} (f : Fin n  A)  ∃[ z ]  i  f i ≲ z

then n = 0 ensures that A is inhabited... and n = 2 gives us the binary definition... hmmm.

UPDATED: but introducing such a definition causes a dependency cycle between

  • Relation.Binary.Definitions
  • Relation.Binary.PropositionalEquality.Core
  • Data.Fin.Base
    grrr....

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Directed definitely does need to include the point! The unbiased definition is cute, but ends up not being the most ergonomic definition. It's akin to defining Monoid via foldMap: this is a useful theorem, but kind of an annoying definition.

For semidirected, we could just call it HasBinaryUpperBounds or something? Also, we will want both UpwardsDirected and DownwardsDirected: we can get them by taking opposites, but experience in agda-categories shows that defining things via duality gives absolutely dreadful goals.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@TOTBWF As I have said elsewhere, in any deployment context in which we want a Directed gadget, adding an explicit inhabitant to the list of hypotheses is/could be seen as a 'neutral' way to resolve this question.

But given the experiments I've made on https://github.com/jamesmckinna/agda-stdlib/tree/directed I'm happy that the Fin-based version seems to have smoother consequences, including inhabitedness built-in. But adding it without further thought/refactoring leads, unfortunately, to a dependency cycle.

Reifying 'Inhabited' plus HasUpperBounds' as a record (as in the original PR, which I objected to for other reasons) may be the solution you seek, but for @JacquesCarette 's (implied) request to avoid a merely binary version (with the need for appeals to Transitiveto go beyondn = 2` in a unifiorm way) in favour of a (more) evenhandedly finitary definition.

I guess things will bottom out when DCPO/Filter/... reaches a stable point, at which point I'd be happy to refactor the original content here, namely that passage to the lifting-with-bottom-element preserve such any directedness, to such a future.

Directed _≤_ = ∀ x y → ∃[ z ] x ≤ z × y ≤ z

-- Generalised connex - at least one of the two relations holds.

Connex : REL A B ℓ₁ → REL B A ℓ₂ → Set _
Expand Down